alexandreteles/bonito-v1-gguf overview

{
  "metadata": {},
  "card_data": {
    "datasets": [
      "BatsResearch/ctga-v1"
    ],
    "language": [
      "en"
    ],
    "library_name": "transformers",
    "pipeline_tag": "text2text-generation",
    "tags": [
      "data generation"
    ],
    "license": "apache-2.0",
    "frontmatter": {
      "datasets": [
        "BatsResearch/ctga-v1"
      ],
      "language": [
        "en"
      ],
      "library_name": "transformers",
      "pipeline_tag": "text2text-generation",
      "tags": [
        "data generation"
      ],
      "license": "apache-2.0"
    },
    "hero_image_url": "https://raw.githubusercontent.com/BatsResearch/bonito/main/assets/workflow.png",
    "summary": "You can find the original model at BatsResearch/bonito-v1",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\ndatasets:\n- BatsResearch/ctga-v1\nlanguage:\n- en\nlibrary_name: transformers\npipeline_tag: text2text-generation\ntags:\n- data generation\nlicense: apache-2.0\n---\n\n# Bonito-v1 GGUF\n\nYou can find the original model at [BatsResearch/bonito-v1](https://huggingface.co/BatsResearch/bonito-v1)\n\n## Variations\n\n| Name | Quant method | Bits |\n| ---- | ---- | ---- |\n| [bonito-v1_iq4_nl.gguf](https://huggingface.co/alexandreteles/bonito-v1-gguf/blob/main/bonito-v1_iq4_nl.gguf) | IQ4_NL | 4 | 4.16 GB|\n| [bonito-v1_q4_k_m.gguf](https://huggingface.co/alexandreteles/bonito-v1-gguf/blob/main/bonito-v1_q4_k_m.gguf) | Q4_K_M | 4 | 4.37 GB|\n| [bonito-v1_q5_k_2.gguf](https://huggingface.co/alexandreteles/bonito-v1-gguf/blob/main/bonito-v1_q5_k_s.gguf) | Q5_K_S | 5 | 5.00 GB|\n| [bonito-v1_q5_k_m.gguf](https://huggingface.co/alexandreteles/bonito-v1-gguf/blob/main/bonito-v1_q5_k_m.gguf) | Q5_K_M | 5 | 5.13 GB|\n| [bonito-v1_q6_k.gguf](https://huggingface.co/alexandreteles/bonito-v1-gguf/blob/main/bonito-v1_q6_k.gguf) | Q6_K | 6 | 5.94 GB|\n| [bonito-v1_q8_0.gguf](https://huggingface.co/alexandreteles/bonito-v1-gguf/blob/main/bonito-v1_q8_0.gguf) | Q8_0 | 8 | 7.70 GB|\n| [bonito-v1_f16.gguf](https://huggingface.co/alexandreteles/bonito-v1-gguf/blob/main/bonito-v1_f16.gguf) | FP16 | 16 | 14.5 GB|\n\n## Model Card for bonito\n\n<!-- Provide a quick summary of what the model is/does. -->\n\nBonito is an open-source model for conditional task generation: the task of converting unannotated text into task-specific training datasets for instruction tuning. \n\n![Bonito](https://raw.githubusercontent.com/BatsResearch/bonito/main/assets/workflow.png)\n\n## Model Details\n\n### Model Description\n\n<!-- Provide a longer summary of what this model is. -->\n\nBonito can be used to create synthetic instruction tuning datasets to adapt large language models on users' specialized, private data.\nIn our [paper](https://github.com/BatsResearch/bonito), we show that Bonito can be used to adapt both pretrained and instruction tuned models to tasks without any annotations.\n\n- **Developed by:** Nihal V. Nayak, Yiyang Nan, Avi Trost, and Stephen H. Bach\n- **Model type:** MistralForCausalLM\n- **Language(s) (NLP):** English\n- **License:** TBD\n- **Finetuned from model:** `mistralai/Mistral-7B-v0.1`\n\n### Model Sources\n\n<!-- Provide the basic links for the model. -->\n\n- **Repository:** [https://github.com/BatsResearch/bonito](https://github.com/BatsResearch/bonito)\n- **Paper:** Arxiv link\n\n## Uses\n\n<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->\n\n### Direct Use\n\n<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->\nTo easily generate synthetic instruction tuning datasets, we recommend using the [bonito](https://github.com/BatsResearch/bonito) package built using the `transformers` and the `vllm` libraries. \n\n```python\nfrom bonito import Bonito, SamplingParams\nfrom datasets import load_dataset\n\n# Initialize the Bonito model\nbonito = Bonito()\n\n# load dataaset with unannotated text\nunannotated_text = load_dataset(\n    \"BatsResearch/bonito-experiment\",\n    \"unannotated_contract_nli\"\n)[\"train\"].select(range(10))\n\n# Generate synthetic instruction tuning dataset\nsampling_params = SamplingParams(max_tokens=256, top_p=0.95, temperature=0.5, n=1)\nsynthetic_dataset = bonito.generate_tasks(\n    unannotated_text,\n    context_col=\"input\",\n    task_type=\"nli\",\n    sampling_params=sampling_params\n)\n```\n\n\n### Out-of-Scope Use\n\n<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->\n\nOur model is trained to generate the following task types: summarization, sentiment analysis, multiple-choice question answering, extractive question answering, topic classification, natural language inference, question generation, text generation, question answering without choices, paraphrase identification, sentence completion, yes-no question answering, word sense disambiguation, paraphrase generation, textual entailment, and\ncoreference resolution.\nThe model might not produce accurate synthetic tasks beyond these task types.",
    "related_quantizations": []
  },
  "tags": [
    "transformers",
    "gguf",
    "mistral",
    "text-generation",
    "data generation",
    "text2text-generation",
    "en",
    "dataset:BatsResearch/ctga-v1",
    "license:apache-2.0",
    "endpoints_compatible",
    "region:us"
  ],
  "likes": 2,
  "downloads": 205,
  "gated": false,
  "private": false,
  "last_modified": "2024-03-16T22:07:45.000Z",
  "created_at": "2024-02-27T05:26:25.000Z",
  "pipeline_tag": "text-generation",
  "library_name": "transformers"
}

{
  "_id": "65dd7281e731928ad80b3bb0",
  "id": "alexandreteles/bonito-v1-gguf",
  "modelId": "alexandreteles/bonito-v1-gguf",
  "sha": "f593102be1cc6eb48f3a3e1d0dba703093b62c7e",
  "createdAt": "2024-02-27T05:26:25.000Z",
  "lastModified": "2024-03-16T22:07:45.000Z",
  "author": "alexandreteles",
  "downloads": 205,
  "likes": 2,
  "gated": false,
  "private": false,
  "pipeline_tag": "text-generation",
  "library_name": "transformers",
  "siblings_count": 16
}